-
Notifications
You must be signed in to change notification settings - Fork 8
/
reproduce.R
552 lines (446 loc) · 17.6 KB
/
reproduce.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% #
#
# ### USAGE:
#
# ## data.frame
# set.seed(1)
# DF <- data.frame("ID"=sample(LETTERS[1:7], 120, TRUE), "SERIES" = sample(seq(1000, 9999, 200), 120, TRUE), replicate(12, round(rnorm(120), 3)) )
#
# # simple call
# reproduce(DF)
#
#
# # notice the difference between `head` and `rows` (see the last few rows in the sample)
# reproduce(DF, head=12)
# reproduce(DF, rows=12)
#
# reproduce(DF, rows=12, shuffle=TRUE)
#
#
# # sample by column name
# reproduce(DF, cols=c("ID", "X4", "X10"))
#
# # sample by column number. (a single number is interpreted via `seq(cols)` )
# reproduce(DF, cols=c(1, 6, 12))
# reproduce(DF, cols=6)
#
# # `head` and `shuffle` apply only to rows, not to columns
# reproduce(DF, cols=4, rows=3)
# reproduce(DF, cols=4, head=3)
# reproduce(DF, cols=4, shuffle=TRUE)
#
# # `head` superscedes `shuffle` and `rows``
# reproduce(DF, cols=4, head=3, rows=100, shuffle=TRUE)
# reproduce(DF, cols=4, rows=3, shuffle=TRUE)
#
#
# # if the whole object is desired, shortcut call:
# reproduce(DF, whole)
# reproduce(DF, whole=TRUE)
# identical(reproduce(DF, whole), reproduce(DF, whole=TRUE))
#
# ## data.table
# DT <- data.table(DF, key=c("ID", "SERIES"))
#
# # notice that with data.table, the row numbers apply to the sample, not to the original source.
# # However, examine the values of ID and SERIES and observe they are different.
# reproduce(DT, head=12)
# reproduce(DT, rows=12)
# reproduce(DT, rows=12, shuffle=TRUE)
#
#
# reproduce(DT, rows=12, shuffle=TRUE)
# reproduce(DT, cols=4, rows=12, shuffle=TRUE)
#
# # head takes priority
# reproduce(DT, cols=4, rows=12, head=7, shuffle=TRUE)
#
#
# ## OTHER DATA TYPES
#
# ## lists
# myList <- list(x=1:5, "hello", "Abc", 1:7, c("a", "b", "c"), c("last", "element", "in", "the", "list"))
# reproduce(myList)
#
# reproduce(myList, 2)
#
# # Shuffle not implemented on lists. Instead use sample
# set.seed(7)
# reproduce(myList, 2, shuffle=TRUE)
# reproduce(sample(myList, 2), 7, head=5, name="myList")
#
#
# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% #
# -------------------------------------------------------------------------------------------------------------------------------------------- #
# ## Update Oct 2013:
# argument `lines.out` has been added.
# The user can specify the total number of lines that the output should be broken up into
# The function attempts to add line breaks at logical points, but this is purely a buest guess based on regex.
#
# example: reproduce(DF, lines.out=4)
# %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% #
# -------------------------------------------------------------------------------------------------------------------------------------------- #
# ---------------------------------------------------------- #
# #
# FUNCTIONS START HERE #
# #
# ---------------------------------------------------------- #
## You can copy and paste the output, or
## if on Mac OS X, the output will be automaticall copied to the clipboard
## ie, just run the function, then hit CMD+v in your text editor or website.
reproduce <- function(x, rows=10, head=NA, cols=NA, clipboard=TRUE, whole=FALSE
, shuffle=FALSE, name=NA, verbose=!silent, silent=FALSE, pos=1
, lines.out=1) {
# If rows is a single number, then that is *how many* rows will be selected
# If rows is a vector of numbers, then that is *which rows* will be selected
# Same for cols. Note, vlaue can be integer or character
#
# This function is data.table aware in that it handles data.table differently from data.frame
# grab the name of x from the previous env.
mc <- match.call()
if (missing(name)) {
mc2 <- mc[[2]]
while(is.call(mc2))
mc2 <- mc2[[2]]
name <- as.character(mc2)
}
d <- dim(x)
## the data.table package may not be loaded, hence this function might be missing
if (!exists("is.data.table"))
is.data.table <- function(x) inherits(x, "data.table")
# allows for the call `reproduce(x, whole)` and setting whole to TRUE and rows to default.
w <- tolower(if(length(mc) >2) as.character(mc[[3]]) else (NA))
if (!is.na(w) && (w=="whole" || w=="hole" || w=="all")){
whole <- TRUE
rows <- 10
}
## CHECK IF LIST
# if(is.list())
# check that rows & cols do not exceed the dims. If so, scale them back
#------------------------
# check that x has dim
if (!is.null(d)) {
# check if `rows` exceeds rows(x)
if (rows > d[[1]])
rows <- d[[1]]
if (length(d)==2 && whole)
cols <- d[[2]]
# check if `cols` exceeds the number of cols in x (if x has cols)
if (length(d) > 1 && !is.na(cols) && is.numeric(cols) && all(cols >= d[[2]]))
cols <- d[[2]]
## error check
if (is.numeric(cols)) {
if (any(cols > d[[2]])) {
if (showWarnings)
warning ("some values of `cols` are not valid. Don't worry, they'll be ignored.")
cols <- cols[cols < d[[2]]]
}
} else if (is.character(cols)) {
if (!all(cols %in% names(x))) {
if (showWarnings)
warning ("some values of `cols` are not valid. Don't worry, they'll be ignored.")
cols <- cols[cols %in% names(x)]
}
}
## check if there are any valid values in cols
if (!length(cols)) {
if (showWarnings)
warning("It appears that `cols` has *no* valid values. Don't worry, we will sample all columns.")
cols <- d[[2]]
}
}
## SAMPLE THE ROWS
# if x has dimension && user did not flag for the whole thing
# we will take only a sampling
if (!whole && length(d)) {
if (!missing(head) && !is.na(head) && missing(rows)) {
x <- x[1:(min(nrow(x), head)), ]
} else if (length(rows) || d[[1]] > rows) {
# check for human error
if (!missing(head))
warning("Both `head` and `rows` has been selected. `head` will be ignored, only `rows` used.")
# if rows has length, then select those specific rows
if (length(rows) > 1) {
# make sure that rows are all present in x
if(is.character(rows))
rows <- setdiff(rows, names(x))
if (is.numeric(rows))
rows <- rows[rows <= d[[1]]]
x <- x[rows, ]
# rows is a single number, and that is how many rows to select.
} else {
# Sample randomly
if (shuffle) {
x <- x[sample(nrow(x), rows, FALSE), ]
# rows of 1 or 2 messes up the arithmetic equation being used as an index
} else if (rows < 3) {
x <- x[1:rows, ]
# If not shuffling, use 60% of the value of rows to grab from the head and the remaing rows to come form the tail
} else {
he <- ceiling(rows * 0.62)
tl <- rows - he
x <- x[c(1:he, 1+d[[1]]-(tl:1)), ]
}
}
}
}
# select only certian columns, assuming cols specified and x has dimension
if (!missing(cols) && !is.na(cols) && length(d)) {
if (is.numeric(cols) && length(cols)==1) {
if (cols > d[[2]])
cols <- d[[2]]
cols <- seq(cols)
}
if (is.data.table(x)) {
x <- x[, cols, with=FALSE]
} else if (length(d) > 2) {
indx <- as.list(c(NA, TRUE, NA, rep(TRUE, d[[2]] - 2)))
indx[[3]] <- cols
indx[[1]] <- x3
x <- do.call(`[`, indx)
} else {
x <- x[, cols]
}
}
## Sample from single dim'd objects
if (!length(d) && length(x)>1) {
L <- length(x)
if (!missing(head)) {
x <- x[seq(min(L, head))]
} else if (!missing(rows)) {
# if rows has length, then take those specific elements
if (length(rows)) {
# make sure that "rows" are all present in x
if(is.character(rows))
rows <- setdiff(rows, names(x))
if (is.numeric(rows))
rows <- rows[rows <= L]
# then sample
x <- x[rows]
# .. otherwise take the first rows-many elements
} else {
rows <- min(rows, L) # make sure rows does not exceed length(x)
# Sample randomly
if (shuffle) {
x <- x[sample(L, rows, FALSE), ]
# sample head and tail
} else {
he <- ceiling(rows * 0.62)
tl <- rows - he
x <- x[c(1:he, (L+1)-(tl:1)), ]
}
}
}
}
## GRAB THE DPUT
z <- capture.output(dput(x))
## remove white space, just because it takes up room
z <- gsub("^\\s+|\\s+$", "", z)
# collapse it all into a single string
z <- paste(z, collapse="")
# data.table:
if (is.data.table(x)) {
# grab the key if there is one, so as to preserve it in the output
k <- key(x)
ky <- if (is.null(k)) "" else paste0(", key='", paste(k, collapse=","), "'")
# enclose in data.table
z <- paste0("data.table(", z, ky, ")")
}
## this is the final output
ret <- paste0(name, " <- ", z)
# remove pointer info (specific to data.table, but might be caught in a list)
pattern <- ", \\.internal\\.selfref \\= <pointer\\: [a-zA-Z0-9]{8,12}>"
ret <- gsub(pattern, "", ret)
## --- SPLIT THE OUTPUT INTO MULTIPLE LINES ---- ##
## Chops the output into lines. Attempts to break lines at reasonable points
## First looks for names(x), if available. Otherwise uses regex.
## Also indents subsequent lines to match up after the `<-`, assuming the data name
## is less than 32 chars long
if (lines.out > 1 && nchar(ret) > 125) {
nc.start <- 2+regexpr("<\\-", ret)[[1]]
if (nc.start > 35)
nc.start > 0
nc <- nchar(ret) - nc.start
widths <- (nc / lines.out)
# where to split the output
splitOn <- if (!is.null(names(x))) regOr(paste0(" ", names(x))) else "\\), \\w+ = "
splitOn2 <- "\\s+" ## basic whiteSpace. Dangerous, since might be in between spaces.
for (i in seq(lines.out-1)) {
prev <- ret[[i]]
ret[i:(i+1)] <- chopLine(ret[i], width=widths, flex=widths+20, splitOn=splitOn, maxNumberOfBreaks=1L, collapse=NULL, showWarnings=FALSE, padToSecondSpace=FALSE, dropTrimmedSpace=FALSE)
## if not chopped, try `splitOn2`
if (identical(prev, ret[[i]]))
ret[i:(i+1)] <- chopLine(ret[i], width=widths, flex=widths+20, splitOn=splitOn2, maxNumberOfBreaks=1L, collapse=NULL, showWarnings=FALSE, padToSecondSpace=FALSE, dropTrimmedSpace=FALSE)
if (identical(prev, ret[[i+1]]))
ret[[i+1]] <- ""
}
ret[2:length(ret)] <- paste0(pasteR(" ", nc.start), ret[2:length(ret)])
ret <- paste(ret, collapse="\n")
}
# if on Mac OSX and flagged to true, then copy to clipboard
if(clipboard && Sys.info()[['sysname']] == "Darwin") {
con <- pipe("pbcopy", "w")
writeLines(ret, con)
close(con)
}
if (verbose) {
cat("\n\n\n\nThis is what the sample looks like: \n\n")
print(x, quote=FALSE)
cat("\n\n")
ln <- function() cat(" ==X", rep("=", 62), "X==\n", sep="")
ln()
cat({if (clipboard) "" else rep(" ", 18)}, sep="")
cat(" Copy+Paste this part.", ifelse(clipboard, "(If on a Mac, it is already copied!)\n", "\n"))
ln()
cat("\n", ret, "\n\n")
ln()
}
# also return invisibly
return(invisible(name))
}
# ------------------------------------------------------ #
# ---------------------------------------------------------- #
# #
# UTILS & SUPPORT FUNCTIONS #
# #
# ---------------------------------------------------------- #
## These functions are normally kept in another file.
# Hence only source them from this file if they are absent
# ie, before every function there is a `if(!exists(..))`
if (!exists("chopLine")) {
chopLine <- function(line, collapse="\n", width=88L, flex=15L
, padding=0L, maxNumberOfBreaks=2L, dotsBeyondMax=FALSE
, padToSecondSpace=FALSE
, trimSpace=TRUE
, showWarnings=TRUE
, splitOn="\\s+", dropTrimmedSpace=TRUE) {
## Takes a line of text and breaks it up into width
if (!is.atomic(line) || length(line) > 1)
stop ("`line` should be an atomic vector of length exactly 1.")
if (!length(line) || nchar(line) < width)
return(line)
## if the split value is not present within the range, return the original value
if (!grepl(splitOn, substr(line, width-flex, width+flex))) {
if (showWarnings) {
warning("Could not find pattern '", splitOn, "within the designated character range (", width , " +/- ", flex," chars)\n")
}
return(line)
}
if (padToSecondSpace) {
if (is.character(padding)) {
warning("`padding` given explicitly as a string, but `padToSecondSpace` is set to TRUE.\nThis will overwrite `padding`.")
padding <- 0
}
# find the first letter of the second word. Then go back one space from there. (We want the END of the second space)
secondspace <- attr(regexpr("(\\s*)([^\\s]+)(\\s+)([^\\s])", line, perl=TRUE), "match.length")[[1]] - 1
padding <- max(secondspace, padding)
}
if (is.numeric(padding))
padding <- pasteR(" ", padding)
## NOTE dont pre allocate. Otherwise would then need to check for blank lines.
ret <- line
# dont chop if maxNumberOfBreaks==0
if (maxNumberOfBreaks)
for (i in seq.int(maxNumberOfBreaks)) {
## only continue if enough chars
if (nchar(ret[[i]]) < width + (.6*flex))
break
spaces <- findCharBetween(ret[[i]], char.to.find=splitOn, from=width-(3*flex), to=width+(4*flex))
if (!length(spaces))
bk <- width
else
bk <- spaces[which.min(abs(spaces - width))]
current <- substr(ret[[i]], 1, bk-1)
pushIt <- ifelse(dropTrimmedSpace && splitOn=="\\s+", 1, 0) # if not splitting on space, do not remove it
new <- substr(ret[[i]], bk+pushIt, nchar(ret[[i]]))
# add a hyphen if mid-work break
if (!length(spaces))
current <- paste0(current, "-")
# clean whitespace
if (trimSpace)
new <- gsub("^\\s+", "", new)
ret[[i]] <- current
ret[[i+1]] <- paste0(padding, new)
}
## add dots
L <- length(ret)
if (dotsBeyondMax && nchar(ret[[L]]) > width) {
ret[[L]] <- paste0(substr(ret[[L]], 1, width-3), " ...")
}
return(paste(ret, collapse=collapse))
}
findCharBetween <- function(string, char.to.find=" ", from=1, to=nchar(string)) {
if (length(string) > 1 || length(char.to.find) > 1 || length(to) >1 || length(from) >1)
stop ("all arguments must be length 1")
if (!length(string) || !length(char.to.find))
stop ("No valid string submitted")
found <- gregexpr(char.to.find, string)[[1]]
inds <- found > from & found < to
found[inds]
}
findFirstCharBetween <- function(string, char.to.find=" ", from=1, to=nchar(string), findLAST=FALSE) {
## TODO: Incorporate findCharBetween
if (length(string) > 1 || length(char.to.find) > 1 || length(to) >1 || length(from) >1)
stop ("all arguments must be length 1")
if (!length(string) || !length(char.to.find))
stop ("No valid string submitted")
found <- gregexpr(char.to.find, string)[[1]]
inds <- found > from & found < to
if (findLAST)
max(found[inds])
else
min(found[inds])
}
}
if (!exists("pasteR")) {
pasteR <- function(x="-", n) {
## Repeats x n-times in a flat string
## When n is non-positive, a blank string, "", is returned
## allow for `pasteR(n)`
if (missing(n) && is.numeric(x)) {
n <- x
x <- "-"
}
# if n is not a single number, iterate
if (length(n) > 1) {
if (length(n) == length(x))
return(mapply(pasteR, x, n))
return( sapply(n, function(n1) pasteR(x, n1)) )
}
# use the chararacter length of n, if it is a string
if(is.character(n))
n <- nchar(n)
# Negative values are considered 0.
n[n<0] <- 0
# otehrwise, simple return
paste(rep(unlist(x), n), collapse="")
}
}
if (!exists("regOr")) {
regOr <- function(vec, brackets=TRUE, asterisk=NULL, escape=FALSE, whole=FALSE, whitespace=FALSE) {
# combines a string vector into a regex `or` statement
if (whole && !brackets) {
warning("Cannot use whole without brackets. Please add `^` `$` anchors manually")
}
if (escape) {
vec <- glob2rx(vec)
vec <- substr(vec, 2, nchar(vec)-1)
vec <- gsub("\\)", "\\\\)", vec)
}
# ws will be pasted. If not grabbing whitespace, make this NULL
ws <- if (whitespace) "\\s*" else NULL
# The main string to return
ret <- paste0(ws, c(vec), ws, collapse="|")
# If not adding brackets, then we're all set.
if (!brackets)
return(paste0(ret, asterisk))
# else return
ret <- paste0("(", ret, ")", asterisk, collapse="")
if (whole) {
# Two separate lines to combat vectorization of paste
ret <- paste0("^", ret)
ret <- paste0(ret, "$")
}
return(ret)
}
}